Trial Discontinuation, Including Unknown Status, and Non-Publication in Randomized Trials of Non-Specific Low Back Pain: A Cross-Sectional Study
Year: 2026
Type: article
Study design: Observational Study
Abstract: BACKGROUND Trial discontinuation and non-publication reduce research value, but field-specific estimates are limited for non-specific low back pain (NS-LBP). We quantified registry-defined discontinuation and non-publication, then explored trial characteristics associated with publication. MATERIAL AND METHODS This primarily descriptive cross-sectional study assessed randomized trials involving adults with NS-LBP that were first posted between 2000 and 2021 and had registered completion dates through December 31, 2023. Registry records were updated through September 1, 2025. Two reviewers independently determined completion and publication status; publication searches continued through January 31, 2026. Discontinuation included terminated, withdrawn, or unknown status. Five prespecified variables were entered simultaneously into an exploratory multivariable logistic regression model. RESULTS Among 433 trials, 112 (25.9%) met the registry-defined discontinuation criterion; 72 of these 112 (64.3%) had unknown status. After excluding unknown-status records, 40 of 361 trials (11.1%) were confirmed as terminated or withdrawn. Among 321 completed trials, 116 (36.1%) lacked a journal publication. Exploratory regression analysis revealed that publication was less likely for multicenter trials (adjusted odds ratio [OR], 0.42; 95% confidence interval [CI], 0.19-0.93) and more likely for trials with enrollment of at least 200 participants (adjusted OR, 3.86; 95% CI, 1.66-8.97), industry funding (adjusted OR, 2.77; 95% CI, 1.25-6.12), and a principal investigator in North America or Europe (adjusted OR, 3.22; 95% CI, 1.92-5.40). Registration period was not associated with publication. CONCLUSIONS Registry-defined discontinuation and lack of journal publication were common, although the discontinuation estimate strongly depended on unknown-status records. Adjusted associations were exploratory and should not be interpreted as causal.
Source: Medical Science Monitor
DOI: 10.12659/msm.954686
Authors: Runqi Wang, Jing Niu, Yang Yang, Dan Wu, Yaping Yu
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7225693878
